UKH

I saw this graph on Facebook and thought it was interesting. Covid and the 'Honnold' effect have led to a hockey stick increase in soloing deaths in America. 

Stats from the AAJ, via Escalade Bliss Climbing on fb. 

I've had this argument on FB already... From memory it doesn't really say anything useful about the UK, as the soloing many of us do here is very different. 

[1] "soloing deaths" should be "severe injuries or deaths during free soloing" to match what's stated in the chart.

[2] "have led to" should be "correlate with". The chart shows correlation, not causation. To get to causation, we'd need to understand more about the circumstances around each of the individual accidents, and also if there have been any other changes that could affect the data (e.g. differences in data collection and record keeping in the 40 years between 1983 and 2023).

That said, it's a pretty reasonable and unsurprising hypothesis.

My first thought would be that climbing has got a lot more popular in the same time . I wonder if the rise in free solo accidents correlate with other stats, like number of outdoor climbers, or number of bouldering or trad accidents.

Given that it's severe injuries as well as deaths, the numbers per year will vary with reporting too. The actual number could be higher, or lower. 

These are tiny numbers. I wouldn't be surprised if emphasis and films had an effect but I think causation is a massive stretch.

In reply to:

Is the '89 spike the 'Croft Effect' then?

It seems to me that soloing harder/longer routes is something that has come in and out of vogue a few times. My impression is that there was a lot of it about in the 70s and 80s then although people obviously carried on dong it it didn't seem to get the same coverage until Free Solo.

In reply to:

What would be the best way you analyse this data? Poisson distribution?

That graphic and annotation doesn't say anything useful about the USA either beyond the raw numbers having gone up.

Maybe it was published with suitable analysis and stated any assumptions in that analysis, but without all that there's not a lot more that can be said.

Firstly, you want to plot it against the number of people climbing full stop. Then see the number of people actually free soloing, then you decide the sample sizes are way too small to draw any conclusions from.

I really am not an expert and I'd like to learn from this but is this not exactly what the Poisson distribution is for? Calculating the probability of an event happening a certain number of times over a given time frame and particularly where the individual probability is low.

Somebody educate me! 

Yes, you could model the rate at which the accidents happen as a Poisson process. A change-point analysis (regression) would be one way to attempt to identify when a change may have occurred. As pointed out above, you would really want to account for any changes in the number of people climbing over time, or better the person hours spent climbing, as an offset term in the regression -- good luck finding that data!

I can't see that there's anything meaningful to be gained from such small figures. 

For instance, in a country of some 300 million people, half a dozen pissed up beginner climbers (e.g. teenage, male) messing around in a quarry, could send things ballistic. 

Mick 

I assume that all you could draw from it is that with a base rate of 0.8pa a year with 7 is very improbable? 

Those were the days!  

I vaguely remember they are either suggesting people learn to rope solo, and possibly offering courses on how to rope solo...

Love it! 

As I wrote the post above, there was an inner voice muttering, "You hypocritical old bastard."

Mick 

As someone else has mentioned, with any kind of analysis, this sample size is way too small. You could have a couple of consecutive years where maybe one particular small community got together and egged each other on to some unhealthy levels of competition. Maybe there was a group on one route and the leader fell and took the rest out. Who knows. You could easily end up with a hockey stick looking graph that isn't indicative of any trend at all. 

Which a sample size this small, the best thing to do would be to look into each individual case in detail and find out what the contributing factors were, rather than starting with a hypothesis and trying to fit it to the sparse data. 

It's good to see that, despite the growth in climbing, absolute numbers of serious injury or death from soloing are still tiny. Compare that to gun deaths in the US (2021: 549 accidental gun deaths... 537 shot dead by police... 458 unknown... and 21,000 murdered, 26,000 suicides*) and you really start to wonder if soloing is safer than going to school in the US.

*  source: https://www.pewresearch.org/short-reads/2023/04/26/what-the-data-says-about...

There's also the hypothesis to be considered. In this case it's tempting to think the hypothesis to have been "Did the release of Free Solo lead to a change in the rate of serious free solo accidents?", whereas in all likelihood the hypothesis came after the data, such as "Is the rise in serious free solo accidents at around the time of the release of Free Solo a result of it?" The choice of hypothesis will substantially change the level of significance required.

Well obviously there's no risk that they're presenting something in a way that might drum up extra business 🤣

[eye roll] 

This doesn't show that Covid and the "Honnold Effect" whatever that is, have had any effect on increased in soloing deaths in America.  It shows the have risen.  At the same time as these other things have taken place, but give the lack of control group it doesn't establish any cause & effect, nor does it subject these figures to the type of statistical analysis to establish whether this is a significant rise, or could be attributed to chance.

There are far too many unmeasured, uncontrolled variables at play to establish any type of causality.

This is very bad science.  

I've just been checking worldwide piracy rates since the release of Disney's "Pirates of the Caribbean" back in 2003.

All I'll say is that ferry ride from Holyhead to Dublin this Summer is going to be a very anxious time for me personally. 

These "stats" would be promoted by escalade bliss. They conveniently run a rope solo course... The fb following is cult-like and a bit weird.

I recall a discussion on here a while back, where somebody asked (something like) how many people had soloed above VS and higher than 30 feet. I was very much in the majority, having done both separately but not both together

That's an interesting combo actually. Lots of people will solo long easy routes and lots of people will regularly solo above VS but the two together is a whole other ball game. I have done both at the same time but not often and not for a long time! 

I would hypothesise that this increase just relates to the increase in climbing participation (and therefore at much smaller numbers, soloing participation) more generally. To relate it to Free Solo sounds like a stretch. It will be interesting to see whether numbers stay higher, or whether this is a short term blip.

I agree, but maybe make single/multi pitch the key distinction. 30 feet is short even for grit!

It's probably self limiting!

It's a gross generalisation but 30ft is probably a fair estimate for how high a climb might be that you might be able to kid yourself is fall-offable without necessarily ending up badly crunched. Especially since most 30ft routes will have a crux which is far lower. It's the crux height and the landing that's really key. If you know for sure that a fall from one of the route's hardest moves will be terminal, or not far off, then to my mind it's true soloing. Below that height and you're really in a hybrid highball zone. I happen to think that doing hard moves in that hybrid zone is much harder than doing them several pitches up, but I don't think that view is typical.

Did you happen to notice whether pet shops had reported any corresponding surge in parrot sales? 

I thought pirate numbers were decreasing and that's the real cause of global warming? There's even a graph to prove it:

https://en.m.wikipedia.org/wiki/Flying_Spaghetti_Monster#/media/File%3APira...

For more of that ilk this website can provide endless examples of things that happen to correlate:

https://www.tylervigen.com/spurious-correlations

On short routes, yes I agree.

"If you know for sure that a fall from one of the route's hardest moves will be terminal, or not far off, then to my mind it's true soloing."

Yep. I think for many, other factors will come into play also on longer routes. Both the subjective feeling of exposure - much more pronounced somehow without a rope - and the ease of escape.

"Below that height and you're really in a hybrid highball zone. I happen to think that doing hard moves in that hybrid zone is much harder than doing them several pitches up, but I don't think that view is typical."

I assume you don't mean the exact same difficulty of moves? That would be very unusual indeed.

It's still just as much soloing however exposed it might be and even if it might be easy to escape. It's just easier to solo when the level of commitment is lower.

That's exactly what I mean, yes. Clearly there's a height below which having the ground nearby makes it easier. Most highballs fall into this category. But there's a grey area just above that where the outcome of a fall would be very hard to predict - probably nasty but not necesarily so. It is this kind of height that I find more difficult than true soloing, as trying to weigh up the consequences of a fall are a distraction from the concentration that, to me, soloing requires.

Totally agree. Find myself weighing things up, rather than getting on with it. 

Above a certain height, barring miracles, you know full well what's going to happen. So there's no more weighing up to do, no more distractions. Consequently I find it easier to be in a bubble of concentration. 

Mick

Yeah I get that.

Back to the OP - nothing wrong with more people dying from falling of climbs as long as they were enjoying themselves.   Put another way - these are the type of twisted stats that will climbing banned (or subject to an HSE inspection) in order to 'save lives'

That combo really needs the VS bit to be above 30 feet. There are plenty of VS's longer than 30 feet which only have easier climbing above 30 feet.

Exactly. That's one of the reasons why a 15m HVS, proper soloing for sure, could be a regular solo for some but the same standard of route in the mountains wouldn't even be considered.

That makes some sense, though as you say, almost certainly isn't typical. I think for many, even in the case of a pretty bad fall, some chance of survival makes doing the moves easier. Consciously it doesn't make much sense as the fall could be horrific, but subconsciously it does.

Don't use FB so no idea what cults go on there, but bliss climbing YouTube channel has some really useful resources there.

Not sure why it was a revelation as (like many good simple ideas) it's obvious with hindsight, but the tip to put a slip knot on runner krabs here and there (any fall will undo the knot, but weight of rope underneath will be taken by runner) was a game changer for me. (Stops the rope weight feed all the rope through to the next catastrophe knot when you get 15 or 20m or something up a pitch)

Those resources could encourage more rope solo (inherently not easy-safe environment) and might increase accidents slightly - then again it might reduce accidents by showing tips to avoid problems.

On balance I'd say they probably reduce accidents?

Let's just blame Honnold though. Working  for a few big companies taught me you should always pick blame first, then investigate to get evidence to support that 🤣

I wonder how many rope solo as opposed to free solo? It just seems like such a faff, the avoidance thereof being one of the key attractions of soloing.

It could actually be down to the release of Spider-Man 

https://www.imdb.com/title/tt2250912

Rope solo (multi pitch) is a *** giant faff on a stick, and a very big stick at that 🤣

Soloing looks pretty safe to be fair. 

By UK you mean the grit? 

It's has to be one of the most heavily climbed on rock types in the UK being where it is and, because many grit crags are small, lends itself to soloing. How many people have soloed the odd easier route Stanage compared to say something on Slate or at Gogarth or Swanage? (Actually I did find one sector at one of the Swanage crags I've visited where I soloed a bunch of easy routes, but it was quite grit like in that it was short and had a flat solid top out.)

My first thought was that the really big spike in growth looks to have occurred in 2021... must be because of climbing being in the olympics 😉